Scheffler Immo E. Mitochondria

Подождите немного. Документ загружается.

yielding highly puriﬁ ed, intact mitochondria was an important contribution.

Such isolated organelles could be viewed by electron microscopy and shown

to have a morphology similar to those viewed in situ .

A plausible explanation made already at the time was that this topology

vastly increased the surface area of the inner membrane. The full signiﬁ cance

of the compartmentalization into an intermembrane space and a matrix could

not be fully appreciated until later, but in this chapter, attention will be largely

conﬁ ned to a discussion of observable morphology of individual mitochondria.

It is evident that the size and shape of mitochondria seen in the electron

microscope depends on the plane of sectioning relative to their long axis.

As more examples of mitochondria in different tissues were examined, dis-

tinguishing features soon became apparent. The matrix was not homogeneous

and exhibited a “ ﬁ ne granularity, ” and frequently there were distinct crystalline

inclusions or small particles of high electron density (Figure 3.3 ). Their number

Figure 3.1 Mitochondrion from centroacinar cell of pancreas ( × 190,000). (Courtesy

of Dr. G. Palade.)

STRUCTURE AND MORPHOLOGY 21

22 STRUCTURE AND MORPHOLOGY. INTEGRATION INTO THE CELL

could in some instances be shown depend on the metabolic state of the tissue

observed. The high electron density may be the result of a sequestration and

storage of calcium. Other studies have suggested that the electron density is the

consequence of the retention of osmium tetroxide by a phospholipoprotein

that may bind calcium in vitro , but may not have such a role in vivo .

Most notable was the variable appearance of the cristae (Figure 3.4 ). They

were often lamellar in appearance, but also tube - like in other cells, and a

Figure 3.2 Schematic view of mitochondrion showing the outer and inner membranes

and the folded cristae. (Drawing provided by Dr. D. Fawcett.)

Figure 3.3 Mitochondrion showing electron - dense inclusions. (Original photograph

of K. Porter; from The Cell , by D. W. Fawcett, 1981, W. B. Saunders Co.)

Figure 3.4 (A) Mitochondria with angular cristae from ventricular cardiac muscle.

(Original photograph of J. P. Revel; from The Cell , by D. W. Fawcett, 1981, W. B. Saun-

ders Co.) (B) Mitochondria with tubular cristae in amoeba. (Original photograph of

T. Pollard; from The Cell , by D. W. Fawcett, 1981, W. B. Saunders Co.) (C) Mitochondria

from adrenal cortex. (Original photograph of D. Friend; from The Cell , by D. W.

Fawcett, 1981, W. B. Saunders Co.) (D) Mitochondrion from astrocyte. (Original pho-

tograph of K. Blinziger; from The Cell , by D. W. Fawcett, 1981, W. B. Saunders Co.) (E)

Mitochondria from ﬁ sh pseudobranch. (Original photograph of J. Harb and I. Cope-

land; from The Cell , by D. W. Fawcett, 1981, W. B. Saunders Co.) (F) Mitochondria from

adrenal cortex. (Steroid secreting cells). (Original photograph of D. Fawcett; from The

Cell , by D. W. Fawcett, 1981, W. B. Saunders Co.)

STRUCTURE AND MORPHOLOGY

24 STRUCTURE AND MORPHOLOGY. INTEGRATION INTO THE CELL

Figure 3.4 (Continued)

complete description required many images, and ideally, serial sections. Thin

sections in the electron microscope still require an interpretation in three

dimensions, and frequently the continuity of a membrane is not seen in a given

section. Thus, tubes become “ free ” circular (vesicular) membranes in cross

section, and sheet - like foldings of the inner membrane in some sections may

appear as ﬂ attened sacs ﬂ oating in the matrix. In the current view the inner

membrane is continuous, and there appears to be no functional reason to have

membrane vesicles within the matrix analogous to the thylakoid membranes

(grana) of chloroplasts.

The interpretation of cristae topology from thin - section electron micro-

scopy by Palade and Sjostrand postulated that the cristae are sheet - like

invaginations of the inner membrane into the matrix of the mitochondrion

(Figure 3.2 ). A maximum number of cristae can be accommodated if the sheet -

like membranes are arranged in parallel, but it seems that the models fudge

Figure 3.4 (Continued)

STRUCTURE AND MORPHOLOGY 25

26 STRUCTURE AND MORPHOLOGY. INTEGRATION INTO THE CELL

the issue of what happens to the image if a different plane of sectioning were

presented. The T - shaped junction cannot be continuous around the entire cir-

cumference of the mitochondrion. Recent advances in electron tomography

have made it possible to view complex 3 - D objects like mitochondria in thick

sections and reconstruct a high - resolution 3 - D image. Data are collected by

recording 2 - D images over a wide range of tilt angles, and 3 - D reconstructions

are achieved by computer - assisted imaging. A 3 - D resolution in the range of

5 – 10 nm can be achieved, hence images of the outer and inner membranes of

mitochondria can be viewed at a hitherto unobtainable resolution (5, 6) . Only

a limited number of tissues have been examined by this powerful methodology

(Figure 3.5A,B ).

Of particular interest and signiﬁ cance is the conclusion by Perkins et al. (5)

that cristae consist of a lamellar portion similar to that envisioned before, but

the continuity with the inner boundary membrane (which lies close to the

outer membrane) is achieved by a relatively small number of narrow tube - like

connections. In other words, though continuous, the inner membrane juxta-

posed to the outer membrane and the inner membrane of the lamellar struc-

tures are connected by a narrow “ bottleneck ” referred to as a cristae junction.

Topologically, the intermembrane space is still continuous with the “ inside ” of

the lamellae, but the authors raise the question whether the bottleneck can

constitute a barrier and hence serve to separate two compartments: intracristal

and intermembrane. It may also serve to segregate patches of inner membrane

with differing protein compositions. Studies exploring the implications of these

morphological observations have been challenging (7 – 12) . The tubular struc-

ture of the cristae junctions requires a mechanistic explanation. On the one

hand, theoretical models have considered the local composition of the lipid

bilayer and the packaging of lipids to create the required curvature (13, 14) .

Other speculations have centered on specialized proteins required for main-

taining cristae junctions. Attempts to purify cristae junctions have not yet met

with success. More recently, several proteins have been identiﬁ ed and impli-

cated in the control of cristae formation, but their relationship to cristae junc-

tions is unclear. The inner - membrane protein mitoﬁ lin is a candidate requiring

serious attention (15) . Mitoﬁ lin is localized in the space between the inner

boundary membrane and the outer membranes, where it can assemble into a

large multimeric protein complex. A perhaps more surprising development is

the discovery in yeast that the ATP synthase (complex V), and speciﬁ cally the

subunits e and g play a role in generating mitochondrial cristae morphology.

The dimerization or oligomerization of the ATP synthase appears to be a

critical step in this mechanism (16 – 18) . The appearance of the cristae and

cristae junctions has been shown to depend on the conditions under which the

mitochondria are prepared for tomography (12) . They are likely to be highly

dynamic in vivo , responding to changes in the metabolic state, ADP/ATP

ratios, ion ﬂ uxes, and matrix swelling. There is evidence that in some but not

all cells, inner membranes are being remodeled during apoptosis, possibly as

part of a mechanism for mobilizing cytochrome c release (19, 20) .

Figure 3.5 (A) Mitochondria seen by electron microscope tomography. Cells were

derived from chick CNS. [The original ﬁ gure was provided by G. Perkins, University

of California, San Diego; with permission from Academic Press (Perkins et al. (5) ).]

(B) The 3 - D reconstruction was achieved after the volume was segmented manually

contouring into the regions bounded by the outer, inner, and cristal membranes. [For

details the reader is referred to the original publication (Perkins et al. (5) .] In the

bottom of the ﬁ gure, individual lamellar sections of crista are shown, together with

the tubular portions which connect the disks to the inner boundary membrane. (The

photographs were generously provided by Dr. G. Perkins, University of California, San

Diego; with permission by Academic Press.) See color plates.

STRUCTURE AND MORPHOLOGY

28 STRUCTURE AND MORPHOLOGY. INTEGRATION INTO THE CELL

The suggestion has been made that cristae junctions represent diffusional

barriers between the IMS and the cristal compartment. If the barrier were

complete, a great similarity to chloroplasts would emerge, where the thylakoid

membranes enclose a compartment distinct from the space between the inner

and outer membranes. Such a view might also imply a difference in the

biochemical composition of different domains of the inner membrane. At the

moment, such speculations have limited experimental support, but a promising

beginning has been made by Vogel et al. (21) . These authors show that proteins

are distributed in an uneven, yet not exclusive, manner between the inner

boundary membrane and cristae membranes.

The number and morphology of cristae are likely to reﬂ ect the response of

the mitochondria to the energy demands of the cell. Highly folded, lamellar

cristae with a large surface area are typically found in muscle cells and neurons,

where the respiratory rate is the greatest. These observations clearly raise

questions about the control of cristae formation and the expression of both

nuclear and mitochondrial genes encoding mitochondrial proteins. A telling

example, although perhaps an extreme one, comes from the study of mito-

chondrial morphology in the yeast Saccharomyces cerevisiae . When grown in

the presence of a nonfermentable substrate such as glycerol or ethanol, the

cells are dependent on oxidative phosphorylation, and hence they have mito-

chondria easily recognizable even by the non - expert. When grown in the pres-

ence of glucose, the biogenesis of the inner membrane is severely repressed;

and the morphology, but not the number of the mitochondria, changes dra-

matically. The details of the mechanism of “ glucose repression ” of respiration

will be presented in a later chapter. A similar change in the abundance and

morphology of an intracellular organelle can be observed for peroxisomes —

for example, in the yeast Piccia pastoris , where growth on methanol or fatty

acids induces the appearance of large peroxisomes. Free - living microorgan-

isms are exquisitely attuned to their environment and capable of adjusting and

optimizing their metabolism to the ever - changing conditions.

The question about the number and shape of mitochondria in yeast,

Saccharomyces cerevisiae , took a different turn when a series of serial sections

were analyzed to reveal that in the extreme there may be only one mitochon-

drion per cell, existing as a highly branched, continuous structure. A single

section alone will show what appear to be discrete mitochondria with spherical

or elongated shapes (22, 23) . (See Section 3.2 for further details.)

In multicellular organisms, mitochondria are likely to have a more invari-

able morphology in a given tissue, but different tissues have distinct energy

needs. It may well be that an altered mitochondrial morphology is also related

to functions of mitochondria not strictly linked to respiration. Steroid secreting

cells such as Leydig cells have mitochondria with an unusually wide range of

sizes and often a very distinct array of tubular cristae (24) . Perhaps the most

puzzling display of cristae morphology is one that occurs sporadically (?) in

many tissues of diverse species. These cristae appear to be made of parallel,

hexagonal arrays of tubules with a triangular cross section. This may be an

extreme manifestation of cristae with sharp angulations regularly spaced

along their length, which are found in mitochondria of cardiac cells, among

others (24) .

It is evident that the more subtle aspects of the variations in cristae mor-

phology can generally not be explained in terms of some sweeping generaliza-

tion about surface area and respiratory activity. The inner mitochondrial

membrane is an exceptional membrane with respect to the protein : lipid ratio.

While in most membranes this ratio is close to 50 : 50, conducive to the image

of “ proteins ﬂ oating in a sea of lipids, ” the inner mitochondrial membrane has

a ratio of ∼ 75 : 25, suggestive of relatively densely packed proteins. Researchers

focused on the inner mitochondrial membrane were among the most reluctant

in their acceptance of the “ ﬂ uid mosaic model ” as a model for all biological

membranes (J. Singer, personal communication). However, numerous studies

have provided strong experimental support for the view that the inner mem-

brane is in a ﬂ uid state rather than a solid state (for a review see references

25 and 26 ). Thus, the proteins of this membrane cannot be exclusively respon-

sible for its folding. Note, however, the role of ATP synthase oligomerization

mentioned above (17) . The discovery of supercomplexes formed by complexes

of the electron transport chain (see Section 5.2 ) also requires considerable

rethinking about the complexes ﬂ oating in a sea of lipids.

If the outer membrane has a ﬁ xed surface area and hence can conﬁ ne the

contents to a ﬁ xed volume, the inner membrane with a larger surface area

must necessarily fold to ﬁ t into the available volume. Some order to this

folding may be provided by the periodic attachment of the inner membrane

to the outer membrane. Such rivet - like attachment points have been observed,

and there is now agreement that such attachment points may be made up of

the components that are responsible for the import of peptides made in the

cytosol. They may be formed transiently by the interaction of integral mem-

brane proteins in the outer membrane with others in the inner membrane, thus

aligning two channels through both membranes. Further details on such a

proposed structure and its function in protein import will be presented in a

later chapter.

Formation and maintenance of tubular cristae or cristae with a triangular

cross section presents a greater challenge to the imagination. Theoretical

studies on the conformation of membranes can lead to stimulating specula-

tions (27) , but experimental veriﬁ cation is still a major unachieved goal. The

morphology of membranes such as the plasma membrane and the Golgi mem-

branes is demonstrably determined by interactions with cytoskeletal elements,

but there is no convincing evidence for the existence of such structures within

mitochondria. Nevertheless, our view of the mitochondrial matrix as an amor-

phous soup of soluble enzymes is most likely too simple - minded. Mitoﬁ lin in

the IMS may represent the ﬁ rst in a number of structural proteins required

for organizing the cristae (15) .

In summary, the variation in mitochondrial morphology has been abun-

dantly documented. Three - dimensional reconstructions from serial sections

STRUCTURE AND MORPHOLOGY 29

30 STRUCTURE AND MORPHOLOGY. INTEGRATION INTO THE CELL

and more recently by confocal microscopy (even in live yeast cells) have

removed some of the last ambiguities of interpretation of thin sections used

in electron microscopy. A general correlation of the total area of the inner

membrane (density of cristae) with the capacity for oxidative phosphorylation

is apparent, but some of the more detailed aspects of cristae morphology in

relation to function and activity are still relatively obscure. Even less is known

about the control of this morphology by nuclear genes in the course of devel-

opment and differentiation. As will be discussed in the following section,

mitochondrial shapes and distributions within a cell may be controlled by

extra - mitochondrial, cytoskeletal elements, but the establishment of the par-

ticular, cell - speciﬁ c interior morphology remains mysterious and therefore a

challenge for the future.

3.2 INTEGRATION INTO THE CELL

3.2.1 Distribution in the Cytosol

So far in the discussion it has sufﬁ ced to consider mitochondria ﬂ oating around

in a cell, with highly diffusible molecules entering and exiting the organelle.

This view is certainly too simplistic, and there are now numerous observations

indicating that mitochondria have a nonrandom distribution in cells; they can

either be restrained in preferred positions, or move or be moved to sites where

their output is needed. They are not only not static in their positions within a

cell, but they are also quite changeable in their shape. Individual mitochondria

can undergo shape changes, or shapes and sizes can be altered by processes

of fusion or ﬁ ssion. Thus, some of the dynamics may be due to internal or

intrinsic driving forces. Other movements or dislocations are quite clearly due

to interactions of mitochondria with the cytoskeleton and the possible involve-

ment of molecular motors.

Studies on the dynamics of shape and position of mitochondria began early

in this century before their identity had even been established. Stains were

used, but the effect of these stains on the “ viability ” of these organelles was

often not clear. Later (1950s) the phase contrast microscope opened up new

ways of exploring live cells and the visible subcellular structures whose identity

was, however, not always fully established. When immunoﬂ uorescence tech-

niques were perfected and speciﬁ c antisera against various identiﬁ ed organ-

elles and cytoskeletal elements became available, another signiﬁ cant expansion

of opportunities for the study of mitochondria was achieved. However, the

need for permeabilization of cells also imposed a limitation: Static images had

to be interpreted that could provide only hints of the dynamics in living cells.

The ﬁ rst indications that mitochondria interact with the cytoskeleton were

made in immunoﬂ uorescence studies with distinct labels for mitochondria and

microtubules (28, 29) . The most powerful technology became available in the

1980s when ﬂ uorescence microscopy and the discovery of new speciﬁ c dyes